Information processing systems increasingly utilize reconfigurable virtual resources to meet changing user needs. For example, cloud computing and storage systems implemented using virtual machines have been widely adopted. Such cloud-based systems include, for example, Amazon Web Services (AWS), Google Cloud Platform (GCP), and Microsoft Azure. Nonetheless, many such cloud-based systems include unused capacity at various times. This is true not only for public and private clouds, but also a wide variety of other types of information processing systems, including systems that do not necessarily utilize virtualization infrastructure. For example, private companies, universities or other enterprises may have considerable amounts of spare capacity available at nights or on weekends in their enterprise information processing systems. In many cases, such spare capacity remains unused. This is due at least in part to the failure of conventional practice to provide an adequate mechanism for encouraging utilization of such spare capacity.